[Doc] Update docs to refer to pooling models (#11093)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2024-12-11 13:36:27 +00:00
committed by GitHub
parent 8f10d5e393
commit cad5c0a6ed
14 changed files with 26 additions and 21 deletions
+2 -2
View File
@@ -893,7 +893,7 @@ class EngineArgs:
'--override-pooler-config',
type=PoolerConfig.from_json,
default=None,
help="Override or set the pooling method in the embedding model. "
help="Override or set the pooling method for pooling models. "
"e.g. {\"pooling_type\": \"mean\", \"normalize\": false}.'")
parser.add_argument('--compilation-config',
@@ -1085,7 +1085,7 @@ class EngineArgs:
"setting --max-model-len to a smaller value.", max_model_len)
elif (self.enable_chunked_prefill
and model_config.runner_type == "pooling"):
msg = "Chunked prefill is not supported for embedding models"
msg = "Chunked prefill is not supported for pooling models"
raise ValueError(msg)
+1 -1
View File
@@ -1085,7 +1085,7 @@ class AsyncLLMEngine(EngineClient):
trace_headers: Optional[Mapping[str, str]] = None,
priority: int = 0,
) -> AsyncGenerator[PoolingRequestOutput, None]:
"""Generate outputs for a request from an embedding model.
"""Generate outputs for a request from a pooling model.
Generate outputs for a request. This method is a coroutine. It adds the
request into the waiting queue of the LLMEngine and streams the outputs
+1 -1
View File
@@ -527,7 +527,7 @@ class MQLLMEngineClient(EngineClient):
*,
inputs: Optional[PromptType] = None # DEPRECATED
) -> AsyncGenerator[PoolingRequestOutput, None]:
"""Generate outputs for a request from an embedding model.
"""Generate outputs for a request from a pooling model.
Generate outputs for a request. This method is a coroutine. It adds the
request into the waiting queue of the LLMEngine and streams the outputs
+1 -1
View File
@@ -209,7 +209,7 @@ class EngineClient(ABC):
trace_headers: Optional[Mapping[str, str]] = None,
priority: int = 0,
) -> AsyncGenerator[PoolingRequestOutput, None]:
"""Generate outputs for a request from an embedding model."""
"""Generate outputs for a request from a pooling model."""
...
@abstractmethod